Company profile

Speechmatics Tech Stack

Speech recognition and transcription AI built on self-trained neural models

Software Development Cambridge 51–200 employees Founded 2006 Privately Held

Speechmatics develops automatic speech recognition models and sells them as transcription and speech-understanding APIs. Its stack pairs PyTorch, TensorFlow and Flash Attention with Kubernetes, Terraform, ArgoCD and GitOps across AWS, Azure and GCP, which points to in-house large-model training rather than reliance on third-party models. Hiring is accelerating and engineering-heavy, centred on senior and principal ML and infrastructure talent.

Tech Stack 39 technologies

Core StackReact JavaScript Mixpanel TypeScript Python TensorFlow PyTorch Docker Kubernetes Ansible Terraform Helm VMware Salesforce Nginx AWS Datadog ArgoCD GitLab OpenTelemetry PostgreSQL Power Query Power BI SvelteKit LinkedIn Sales Navigator Flash Attention Azure cURL GCP GitOps+9 more

What Speechmatics Is BuildingChallenges and projects from their job posts

◆Challenges

Optimising model speed, accuracy and cost
Removing friction across the ML lifecycle
Infrastructure stability and disaster recovery testing
Driving sales pipeline growth
Reducing operational toil through automation

▲Active Projects

Training large self-supervised speech models
Production deployment and evaluation infrastructure for models
Scalable training-data pipelines and web data acquisition
Infrastructure strategy and capacity planning
Custom API integrations and developer demos

Hiring Activity

Accelerating10 roles · 5 in 30d

Department

Engineering

Sales

Seniority

Senior

Lead

Mid

Principal

Companies like Speechmatics

4,500+ Software Development companies running React and JavaScript

See the full list

About Speechmatics

Speechmatics is a Cambridge, UK-based speech technology company founded in 2006, with an additional office in New York. It builds machine learning models that convert spoken audio into text, both in real time and from recorded media, across many languages and accents. Businesses build products on top of these transcripts, adding capabilities such as summaries, sentiment, topics and translation. The privately held company has 51–200 employees and hires in the United Kingdom, the United States and Serbia.

HeadquartersCambridge

Company Size51–200 employees

Founded2006

Hiring MarketsUnited Kingdom, United States, Serbia

Frequently Asked Questions

What tech stack does Speechmatics use?

Speechmatics uses Python, PyTorch, TensorFlow and Flash Attention for ML, with Kubernetes, Docker, Terraform, Helm and ArgoCD on AWS, Azure and GCP. React, TypeScript and SvelteKit cover the front end.

Is Speechmatics hiring engineers?

Yes. Hiring is accelerating and mostly in engineering, skewed toward senior, lead and principal roles, with some sales hiring, across the United Kingdom, the United States and Serbia.

What is Speechmatics working on?

Job posts point to training large self-supervised speech models, deploying them into production, building scalable training-data pipelines and building model deployment and evaluation infrastructure.

Similar Companies in Software Development

Other companies in the same industry, closest in size

How this profile is built

Speechmatics's technology stack, projects, and hiring signals are inferred from public hiring and company data — career pages, public listings, and company web presence — then clustered and de-duplicated. Figures are estimates that refresh over time. Read our full methodology →

This is not an official vendor or customer list. It is a technology-adoption signal inferred from public data, intended for B2B research.